[PATCH] Upstream hash: speedup consistent hash init

[Roman Arutyunyan]

Hello!

On Thu, Feb 19, 2015 at 01:33:03PM +0800, Wai Keen Woon wrote:
> On 2/18/2015 8:49 PM, Roman Arutyunyan wrote:
> >>Note that in the original implementation, if there are points
> >>with duplicate hash, only the first is kept. In this change, all
> >>are included.
> >This is the intended behaviour.  Consistent hash array is build over
> >server entries, but not addresses resolved from them.  Duplicate points
> >are ignored since most likely they refer to multiple addresses of
> >the same host.
> I see. I could add a loop to remove duplicate hashes and maybe adjacent
> points referencing the same server too. Or do you prefer to take some time
> to look into it in more detail first?

Thanks for your work.

We have committed a slightly modified version of the patch.

-- 
Roman Arutyunyan